Nutrition scientists say research to support nutritional health claims should not be limited by randomized controlled trials – the so-called gold standard of evidence that has yielded disappointing results linking nutrients to health benefits.

Rather, researchers and regulators should consider measuring the physiological effects of nutrients in the body or simply accepting a lower degree of certainty for related health benefit claims, given the low risk profile of most nutrients, argued presenters at the Institute of Food Technologists’ annual Wellness conference March 28 in Rosemont, Ill.

Robert Heaney, a professor of medicine at Creighton University, said FDA and other regulators should consider adopting a “least-harm strategy” to consider whether permitting a health claim for a nutrient would be more or less harmful than prohibiting it and risking a population of nutrient-deficient consumers.

Jeffrey Blumberg

Photo courtesy of Tufts University

Heaney, who specializes in the study of osteoporosis, vitamin D and calcium physiology, added that a problem with many RCTs looking at essential nutrients is they fail to consider their subjects’ nutrient intake levels prior to supplementation. But it would be unethical to begin with a control group of deficient subjects and continue depriving them of an essential nutrient, he said.

“Furthermore, if somehow you could twist your conscience so that this might be rationalizable, you’d find it’s unfeasible,” he said, “because the people who would volunteer for this study are health-conscious, and they never would’ve been down to that lower rate to start with.”

Jeffrey Blumberg of Tufts University echoed the sentiments of many dietary supplement industry stakeholders in pointing out that RCTs are best suited to measure drug treatment efficacy and toxicity, not nutrient function.

Blumberg, director of the Antioxidants Research Laboratory in Tufts’ Friedman School of Nutrition Science and Policy, suggested nutrients, with a much larger gap between efficacy and toxicity compared to pharmaceuticals, should have a lower bar of scientific certainty for supporting health claims.

Otherwise, regulators and scientists “are going to impede [industry’s] ability to develop new food products and give them to people and tell them why they are good for them,” Blumberg said.

Selling FDA and the Federal Trade Commission on a lower degree of certainty for health claims likely would be difficult. Blumberg added that firms marketing products with such claims also would have to conduct education campaigns to help consumers differentiate among varying levels of evidentiary strength for claims.

Without a lower bar of certainty for supporting health claims, regulators and scientists “are going to impede [industry’s] ability to develop new food products and give them to people and tell them why they are good for them.”

– Jeffrey Blumberg, Tufts University Antioxidants Research Laboratory

Representing industry at the IFT session, Council for Responsible Nutrition executive Douglas “Duffy” MacKay noted FDA’s existing paradigm for approved health claims is causing stagnation, with no new approved health claims since 2008 when the agency allowed vitamin D to join calcium in the health claim for reduced osteoporosis risk Also see "FDA Adds Vitamin D To Osteoporosis Risk Reduction Health Claim" - Pink Sheet, 6 October, 2008..

“Industry is sort of giving up,” said MacKay, the trade group’s VP of scientific and regulatory affairs. “They’re not willing to make that investment without the intellectual property protection. So we need improvements in science and technology to advance this field.”

Desperately Seeking Surrogate Endpoints

The search for more suitable frameworks for studying nutrients comes as the scientific community and the nutritional products industry continue to butt heads about the validity of studies showing little to no evidence of nutritional benefits on specific health conditions.

Industry stakeholders went on the defensive after a meta-analysis this month in the Archives of Internal Medicine showed inconclusive evidence of omega-3 fatty acid supplements’ beneficial effects in secondary prevention of cardiovascular disease ([A#05120416003]). Both CRN and the Natural Products Association argued the 14 randomized, double-blind, placebo-controlled trials included in the analysis had been cherry-picked and the researchers omitted other data showing a positive correlation.

Blumberg suggested observational cohort studies with appropriate biomarkers of exposure and response could have more evidentiary value than many regulators and scientists currently allow. Even animal studies, using technological advances such as “humanized mice,” are generating data that increasingly can be extrapolated to humans, he said.

Additionally, new tools are emerging to allow more accurate assessments of nutritional intervention effects, Blumberg said. This includes a “global health index” approach that evaluates multiple systems in the human body, with the understanding that nutrients act in a pan-systemic manner.

In performing RCTs on nutritional interventions, Blumberg noted some researchers are beginning to recognize that nutrients help maintain a body’s physiological processes within a normal range. By interrupting homeostasis in an otherwise healthy individual – through a high-fat meal, for instance – the researchers can perform the nutritional intervention to gauge whether it “can blunt that response” and help restore homeostasis, Blumberg said.

Translating such assessments into health claims will require validated biomarkers and surrogate endpoints, which are scarce at present, MacKay noted.

“Expanding on biomarkers and surrogate endpoints would really fast-forward the industry’s ability to do science.”

– CRN’s Douglas “Duffy” MacKay

While FDA recognizes some validated biomarkers, including bone density as a surrogate for osteoporosis, scientific inertia may be preventing the development of additional biomarkers. For example, MacKay noted a proposed health claim for lutein and reduced risk of macular degeneration failed because increased macular pigment density was not a validated biomarker for macular degeneration.

“Expanding on biomarkers and surrogate endpoints would really fast-forward the industry’s ability to do science,” MacKay said.

The supplement industry hoped in 2010 that FDA would open the floodgates for more validated biomarkers following the Institute of Medicine’s report recommending a framework for assessing biomarkers for chronic disease Also see "IOM Proposal Could Open Door For Biomarkers In Supplement Research" - Pink Sheet, 17 May, 2010.. So far, the agency has not indicated plans to adopt such a framework.

In Defense Of RCTs

While novel technology and approaches may help improve the ability of RCTs to capture nutrient effects, Blumberg said often “the shoe doesn’t fit” and employing an RCT simply is not appropriate or feasible.

But some audience members countered that RCTs still hold an important place in nutritional investigations and should not be dismissed out of hand.

During a question and answer session, nutrition scientist Brian Larson said the discussion surrounding RCTs should focus on the study design rather than the RCT framework itself.

“Random controlled trials can be poorly designed, poorly interpreted and misunderstood like any other type of study design,” Larson, a nutrition consultant with JG Consulting Services in Dowling, Mich., said in a follow-up interview.

“The setup of an RCT is eminently able to discern differences if you design the study appropriately,” he said.

Further, with the majority of the U.S. adult population deficient in essential nutrients such as vitamin D and zinc, Larson sees less of an ethical concern in recruiting study participants when their nutrient levels already are low.

Kathy Musa-Veloso, of consultancy Cantox Health Sciences International, also defended RCTs for studying nutrients, especially in assessing their potential benefits beyond the established recommended daily allowance or adequate intake levels. Such a setup eliminates any ethical issues that come with possibly depriving subjects of essential nutrients.

“If the intent is on studying the nutritional benefits of consuming a nutrient that is several-fold greater than the RDA or AI, then there is no reason why an RCT cannot be conducted,” said Musa-Veloso, associate director of Cantox’s food and nutrition group.

Like Larson, she underscored the importance of study design and, like Blumberg, suggested the need for compromises. In an email, Musa-Veloso said RCTs for nutrients should feature a more forgiving statistical confidence level, given that nutrient effects often are subtle and the placebo group likely will intake incidental amounts of the nutrient being studied.

Nutrition Scientists Float “Least-Harm Strategy” To Allow Health Claims

More from Clinical Trials

More from R&D